1 Acknowledgements Many wonderful people have been a part of this project. I would first like to thank Dr. Maggie Cao. Her wit, scholarly example, and reassurance has been a guiding light throughout the thesis project and my time at UNC. I will never forget when she first encouraged me to search for questions rather than answers. I look forward to more conversations in years to come. Thanks to Dr. Bernie Herman’s special ability to pinpoint the kernels of any muddled truth. His synthesis of knowledge is contagious and I am lucky to have his mentorship. Special thanks is owed to Dr. Cary Levine, whose lectures make subtle appearances throughout this text. Special thanks to Professors Mark Hansen, Stanley Abe, and Daniel Anderson for reading earlier versions of this project. Their willingness to impart knowledge is a testament to mentorship. Unbeknownst to me, this thesis began when I wrote my first research project in the art history department with Dr. Daniel Sherman. A paper on Carol Summer’s screenprint, Kill ​ ​ for Peace, concretized both my interest in art history and my four-year search for holes. Dr. Sherman’s scholarly ​ ​ mentorship, balanced critique, and encouragement had a profound impact on my college experience. I thank him for pushing me towards art history. Thanks is also given to Josh Hockensmith, Alice Whitesell, and all of the library staff at the University of North Carolina. Countless other maintainers of information helped form this thesis throughout the country. The Winterthur Library, United States Postal Museum Library, the National American History Museum, Yale Libraries, and Harvard Libraries all were welcoming and gracious with my questions. Likewise, this projected was enriched by the John and June Alcott award. Thanks to the entire Art History department’s constant support for pushing my thinking. You all made my time at Carolina thoroughly enjoyable. None of this would be possible without the people who shaped me. Thanks to my grandmothers, Alma and Joyce, whose persistent jabs, jeers, and threadedness made me the critically attuned person I am today. Thanks to my father who taught me spontaneity and perseverance. Thank you for being a balanced sounding board for so many of my ideas. Thanks for my mother, a fellow Tarheel, who taught me to be empathetic and keenly judgmental -- two of the most necessary skills for any art historian. To Noble and Zayla: I am expecting big things from both of you! And, finally, to my friends. Thanks for hanging around. 2 Introduction A jpeg of Gilbert Stuart’s famous George Washington portrait (Figure 0.1) has been dragged from Google ​ ​ onto my desktop as “download.jpg.” (Figure 0.2) Clicking on the icon conjures a larger version in Apple’s Preview ​ ​ ​ program, which, beyond viewing the image’s dimension in pixels, allows me to edit the scale, color, and background of Washington’s likeness. The digital image copied from the internet is mine to manipulate. When I open download.jpg as a file in a text editor, however, the image completely dissolves from the screen. Instead, George ​ Washington’s likeness transforms into an indecipherable code; swaths of paint are substituted for lines of letters, numbers, symbols, and punctuation. “ì‡Qú8Â7+E vrÛK ÀÎy߃ñ6VÌw¥iÕ«';aÊ| ˝Øk‚®öyïˇŸ c:1ågR,” anyone? (Figure 0.3)1. Whether made on the computer, uploaded, or scanned, digital images are stored and transmitted across the world’s screens instantaneously. As download.jpg’s coded textuality makes clear, the viewing of digital images ​ ​ does not require the perception of the many layers operating beyond the immediate interface. If anything, digital images naturally instantiate themselves as completely indifferent to the code, labor, and various materials that lie beneath their pixelated surfaces. Images are traditionally phenomenological objects that can only be viewed through our sense of sight; images exist only when the eye is able to perceive them. Digital images, as my example shows, exist in their totality ​ ​ as sets of data, merely coordinate points free from a visual plane. Computer graphics exist, therefore, prior to their instantiation in and as a given image.2 As media theorist Friedrich Kittler notes in his lectures on optical media, “computers must calculate all optical or acoustic data on their own precisely because they are born dimensionless and thus imageless. For this reason, images on computer monitors […] do not reproduce any extant things, surfaces, or spaces at all. They emerge on the surface of the monitor through the application of mathematical systems of equations.”3 While mimetically the same as images portrayed on tactile surfaces, computer graphics operate differently than those of painting, printing, camera, and film. In simulating an image, the very material basis - that 1 For a more radical use of this tooling, see, Hito Steyerl, “Medya: Autonomy of Images” in Duty Free Art. (New York: Verso, ​ ​ ​ 2017) 2 Jacob Gaboury, “Hidden Surface Problems: On the Digital Image as Material Object.” Journal of Visual Culture 14, no. 1, ​ ​ ​ (April 2015), 40–60. 3 Friedrich Kittler, Optical Media : Berlin Lectures 1999, (Cambridge, UK ; Malden, MA: Polity Press, 2010.) ​ ​ ​ 3 which lies beneath the surface of the screen - is largely forgotten. Contrasting early philosophies that equated vision and truth, digital images are the ultimate surface sham. Digital images break the contract of empirical perspective through their virtual simulation. As William Ivins describes in the Rationalization of Sight, perspective is merely a “means for securing a rigorous two-way, or ​ ​ reciprocal, metrical relationship between the shapes of objects as definitely located in space and their representations.”4 Download.jpg is thus simulated on my desktop much like Gilbert Stuart’s Washington is hung on ​ ​ ​ ​ the walls of the Boston Athenaeum, but has a completely different ontological existence in space and time. Download.jpg seems totally opposed to other tactile iterations of Stuart’s portrait. The surface effects of ​ download.jpg’s pictorial mechanisms - its virtual composition, coded textuality, and pictorial becoming - are hidden ​ ​ from sensorial perception, completely removed from the human’s ability to see and know. Moreover, while we can attest that the digital graphic was at one point a painting, the image exists entirely removed from earlier iterations. The viewer cannot immediately tell what camera was used to scan the image, the labor it took to make the image, nor the data necessary to keep it accessible on Google. The surface of digital images betrays perception and the varied history of the image. Out of sight and out of mind, the digital image presents itself as an object for our use without regarding its own history of making. The operands surrounding the material making of download.jpg began long before computers were ​ ​ actualized. Although there was no such thing as a computer image in the mid-nineteenth century, images could, in fact, exist beyond their actualization on a tactile surface. No longer optical, they were digital representations. My choice to download Gilbert Stuart’s Washington was intentional, for the beginning of these screen tactics and the ​ ​ re-negotiation between surfaces and epistemic modes of knowing were brokered upon the making of a peculiar copy of the nation’s forefather. Created in 1851 by a Messrs. Ponson, Philippe, & Vilbert for an American ambassador, ​ ​ the image (Figure 0.4) seems an almost near perfect copy of a print of Stuart’s Washington (Figure 0.5) or perhaps ​ ​ even a Daguerreotype copy of the original painting. The object, however, is not made from paper or canvas or glass. The image does not sit upon its material substrate; the image and object are one and the same. Not painted, not printed, nor photographed -- the image is made of silk threads and was woven upon a Jacquard loom. 4 William Ivins, On the Rationalization of Sight, with an Examination of Three Renaissance Texts on Perspective, (Da Capo ​ ​ ​ Press, 1973.) 4 Invented sometime at the beginning of the nineteenth century by Joseph Marie Jacquard in Lyons, France, the Jacquard Loom revolutionized weaving: simultaneously making the laborious process much faster as well as enabling a new visual definition. The Jacquard loom, or, rather, the Jacquard harness, was a device that automated the raising of warp threads of the loom between each passage of the horizontal weft thread. The automation of the loom was controlled by a system of interlaced pasteboard punch cards. Holes were punched on each card corresponding to a particular alignment of the warp rod, during a single passage of the horizontal shuttle. Once energy was applied, the perforated cards were drawn along a constantly rotating metal box. The cards were bisected and ‘read’ by needles connected to the rods controlling vertical warp threads. The needles that lined up with the card’s punched holes would fall, shifting the corresponding rods and their threads to the “on” position. The other rods, those whose needles bisected the part of the card without any perforation, remained in the “off” position.5 The Jacquard loom was thus a machine for reading and translating optical data through the use of binary code. ​ ​ Historians of the digital world have long acknowledged the Jacquard loom as the beginning of the computer’s history. There is often, however, an impulse to reduce the importance of the Jacquard loom as one of the plethora of technologies that were used to merely create the computer. As the accepted history proceeds, binary punch cards of the loom inspired Ada Lovelace and Charles Babbage to dream up Analytical Engine, which would algebraically “weave patterns as the Jacquard Loom weaves flowers and leaves.”6 Babbage’s Analytical Engine was never built, but was eventually actualized by American Herman Hollerith for the census calculation machine and then used throughout the twentieth century for the International Business Machine Corporation.